Main
Jinseo Choi
I am a CS M.S. student at Gachon university, advised by Prof. Donghyun Kang. I am currently working in Noslab. My research interests lie in computer systems, with a focus on operating systems, storage systems, and virtualization. Recently, I am interested in accelerating machine learning pipeline focused on computer systems. I like technical conversations about any CS field that gives insight. If anyone interested, please contact me.
Education
M.S. student, Computer Engineering
Seongnam-si, Gyeonggi-do
Gachon University
Current - 2023
- Focused on Machine learning Pipeline Optimization.
- Running project to implementation blockchain network on IoT device and stability evaluation.
B.S., Computer Engineering
Changwon-si, Gyeongsangnam-do
Changwon National University
2022 - 2018
- Focused on Linux kernel, Storage system Studied Operating system, Computer network, Software engineering, Alogrithm, Data structure, Database
- Ran project to Designed a housing price index prediction model with local-economic data and developed housing policy establishment assist application for local-government officer.
A.S., Robot Mechatronics
Changwon-si, Gyeongsangnam-do
Masan University
2018 - 2016
- 3D modeling(CATIA), PCB design(OrCAD), Embeded programming(ATmega, ARM Cortex)
Research Experience
Master’s course researcher
Noslab
Gachon University
Current - 2023
- Developed machine learning pipeline parallelism framework between CPU and GPU. Analyzed the PyTorch framework with a focus on GPU launch.
- Ran project to deep learning based marine P&ID digital transformation.
Master’s course researcher
Noslab
Changwon National University
2022 - 2022
- Analyzed TensorFlow with I/O interference. Developed data shuffle by parallelism. Analyzed Machine learning pipeline bottleneck.
Undergraduate Researcher
Noslab
Changwon National University
2022 - 2018
- Analyzed linux kernel I/O scheduler, Cgroups virtualization. Studied storage system(especially SSD). Test DiskSim, SSDSim and legacy code porting to current system.
Publications
딥 러닝 훈련 시간 개선을 위한 쓰레드 기반 셔플링 기법
정보과학회 컴퓨팅의 실제 논문지(KTCP), Vol. 46, No 6, pp. 75-80, 2022.
N/A
2022
- 최진서, 강동현
훈련 및 검증 성능 개선을 위한 텐서플로우 병렬 처리 기법
정보과학회논문지(JOK), Vol. 49. No. 6, pp. 407-415, 2022
N/A
2022
- 최진서, 강동현
딥 러닝 기반 테이블 검출 기법
한국컴퓨터종합학술대회 논문집(KCC), pp. 930-932, 2022.
N/A
2022
- 최진서, 강동현
Overlapped data processing scheme for Accelerating Training and validation in machine learning
IEEE Access, Vol. 10, pp. 72015-72023, 2022
N/A
2022
- 최진서, 강동현
딥 러닝 연산을 위한 GPU/CPU 성능분석
한국컴퓨터종합학술대회 논문집(KCC), pp. 1145-1147, 2022.
N/A
2022
- 최진서, 강동현
I/O 간섭에 의한 텐서플로우 성능분석
In Proceedings of the 한국컴퓨터종합학술대회 논문집(KCC), pp. 1145-1147, 2016.
N/A
2021
- 최진서, 강동현
- 학부생 부문 우수논문상🏆
Interleaved data processing scheme for optimizing tensorflow framework
In Proceedings of the IEEE 11th International Conference on Consumer Electronics(ICCE), pp. 1-3, 2021.
Berlin, Germany
2021
- 최진서, 조민선, 강동현